This proposal aims at training an MD/PhD bioinformatician in the statistical genetics of DNA sequence analysis, and in psychiatric genetics more generally, through a project that takes a novel approach to analyzing next-generation sequencing (NGS) data for detection of rare variants critical to bipolar disorder (BP) susceptibility. BP has a substantial genetic component, but the underlying genetic and neurobiological triggers of the illness remain elusive. While BP genome-wide association studies (GWAS) have focused on common variants and recently identified several SNPs associated with BP, there remains a much-discussed missing heritability problem. This strongly suggests that rare variants might play a critical role in BP susceptibility. With the advent of next-generation sequencing, a relatively low-cost, high-throughput approach to scanning the genome for rare variants is now available. While sequencing technology has seen enormous leaps forward and can now provide us with a torrent of data, the methods for making sense of the data are still catching up. While disease-based sequencing can turn up thousands of rare variants in cases, there exists no robust method for determining which of them plays an etiologic role in the illness. The applicant will take advantage of a unique resource, a newly funded project to use NGS of 800 BP cases and 800 controls across the 1,500 synaptome genes (which encode proteins expressed in the brain synapses), to implement novel methods for analysis of rare susceptibility variants for BP. The applicant proposes to take advantage of a powerful recently developed statistical technique for dealing with rare events, exact logistic regression, and incorporate it into an algorithm for assessing associations between rare variants and disease. This project will be accomplished through the following specific aims: 1) Develop new bioinformatics tools for functional characterization and categorization of identified genetic variants, 2) Develop novel analytic methods to test associations of rare variants with disease, and 3) Test for association of rare variants in BP. This proposal will allow for the development of the applicant as an expert in bringing together bioinformatics and statistical genetics. Results from this study will provide new insights to the etiology and pathophysiology of BP, an important psychiatric illness in the population. In addition, it will provide a set of tools that will be widely applicable to the study of rare variants across complex common diseases.